Toyota Research Institute, Inc.

Toyota Jidosha Kabushiki Kaisha

Leland Stanford Junior University

Massachusetts Institute of Technology

Nikos Arechiga Gonzalez

Jonathan Decastro

Frank Permenter

Soon Ho Kong

Scott Carter

Dennis Park

Soonho Kong

Sumanth Dathathri

Shashank Jivan Vernekar

Karen Y Leung

Daniel Stonier

David Ayman Shamma

Guy Rosman

Sudeep Pillai

Shinichi Shiraishi

Luke S Fletcher

Matthew L Lee

John J Leonard

Evangelos Kokkevis

Marco Pavone

Rumen Iliev

Emily Sarah Sumner

Alexandre Filipowicz

Daniel N Jackson

Francine Chen

Richard Poyner

Sagar Behere

Dimitrios Koutentakis

Mike Meichang Wang

Feng Ping Angela Leong

Xin Zhang

Alexandre Leo Stephen Filipowicz

Daniel D Kang

Systems and methods are provided for implementing soft model assertions (SMA) system and techniques designed to monitor and improve Machine Learning (ML) model quality by to detecting errors within the one or more ML models. SMA techniques and systems are distinctly designed to leverage: 1) a user's ability to specify features over data; and 2) large, existing datasets of organizations, in a manner that can improve the accuracy and quality of predicting potential errors in Machine Learning (ML) models. A SMA system can include a controller device receiving predictions generated based on the ML models and output from the SMA system. The controller performs autonomous operations of the system in response to determining that the one or more detected errors within the one or more ML models yield a high certainty of errors in the predictions. The SMA system also includes a domain specific language and a severity score module.

Systems and methods for soft model assertions

A method for an electric vehicle trip simulator is described. The method includes determining parameters for an upcoming trip of an electric vehicle, including at least a start location and a destination location. The method also includes analyzing data associated with a plurality of points of interest between the start location and the destination location. The method further includes selecting a set of locations from the plurality of points of interest for the upcoming trip of the electric vehicle, according to the analyzing of the data. The method also includes displaying a travel order recommendation associated with the set of locations selected for the upcoming trip of the electric vehicle, including at least one charging station location according to a current charge capacity of the electric vehicle.

Trip simulator for increased time-efficiency when charging

Systems and methods for a powered, robotic exoskeleton, or exosuit, for a user's limbs and body are provided. The exosuit may be equipped with airbag devices mounted at various locations on the suit. The exosuit may include on-board computing equipment that can sense, compute control commands in real-time, and actuate limbs and airbags to restore stability (fall prevention) and minimize injuries due to falls, should they happen (fall protection).

Wearable exoskeleton

Systems and methods for generating and evaluating driving scenarios with varying difficulty levels is provided. The disclosed systems and methods may be used to develop a suite of regression tests that track the progress of an autonomous driving stack. A robustness trace of a temporal logic formula may be computed from an always-eventually fragment using a computation graph. The robustness trace may be approximated by a smoothly differentiable computation graph, which can be implemented in existing machine learning programming frameworks. The systems and methods provided herein may be useful in automatic test case generation for autonomous or semi-autonomous vehicles.

Systems and methods for automatic test generation

A method for certified control of a self-driving ego vehicle is described. The method includes analyzing a safety situation of the self-driving ego vehicle to determine a proposed vehicle control action using a main controller of the self-driving ego vehicle. The method also includes presenting, by the main controller, the proposed vehicle control action to an interlock controller, including a certificate of the proposed vehicle control action. The method further includes checking a safety certification evidence from the certificate by the interlock controller using a predefined safety argument to verify the safety certification evidence of the certificate. The method also includes directing, by a low-level controller, the self-driving ego vehicle to perform a certified vehicle control action.

Certified control for self-driving cars

Systems, methods, and other embodiments described herein relate to informing a user how a hypothetical electric vehicle would perform based on the driving habits of the user in a non-electric vehicle. In one embodiment, a method includes identifying an origin and a destination of a non-electric vehicle after the non-electric vehicle has arrived at the destination. The method includes determining energy information for a hypothetical electric vehicle traveling from the origin to the destination and outputting the energy information to the user.

Electric vehicle simulation

Systems and methods are provided for requirements engineering, and may include: receiving as input, time series data from at least one of a simulation of a vehicle run on a simulation system, or from the vehicle in operation; a requirements monitoring system checking to determine whether a plurality of requirements for operation of the vehicle are met, wherein the requirements are expressed in signal temporal logic form and a requirement includes at least an associated minimal sampling rate and a filtering policy applicable to the requirement; determining a quantitative conformance for each of selected requirements of the plurality of requirements; and add requirements to a verified requirements set based on the qualitative conformance of the requirements.

Systems and methods for monitoring specifications over simulation and test data

A personalized notification method for a mobility as a service (MaaS) vehicle includes receiving conditional personalized notification features from a passenger of the MaaS vehicle. The method also includes monitoring current driving environment of the MaaS vehicle to determine whether a condition of the conditional personalized notification features is satisfied. The method further includes notifying the passenger when the condition is satisfied via at least one localized output device in a compartment of the MaaS vehicle.

Personalized notification system for mobility as a service

A method for synthesizing parameters for control of a closed loop system based on a differentiable simulation model of the closed loop system includes determining requirements/specifications for the closed loop system in signal temporal logic (STL). The method also includes selecting a parametric control law having a differentiable parameter control function. The method also includes converting the requirements in signal temporal logic into differentiable computational graph. The method further includes building the differentiable simulation model as a differentiable computational graph. Furthermore, the method includes automatically learning values of parameters for the differentiable parameter control function of the closed loop system by backpropagating an error.

Simulation-based technique to synthesize controllers that satisfy signal temporal logic specifications

An artificial intelligence perception system for detecting one or more objects includes one or more processors, at least one sensor, and a memory device. The memory device includes an image capture module, an object identifying module, and a logical scaffold module. The image capture module and the object identifying module cause the one or more processors to obtain sensor information of a field of view from a sensor, identify an object within the sensor information, and determine at least one property of the object. The logical scaffold module causes the one or more processors to determine, by a logical scaffold, when the at least one property of the object as determined by the object identifying module is one of a true condition or a false condition.

System and method for detecting errors and improving reliability of perception systems using logical scaffolds

System, methods, and other embodiments described herein relate to modeling dynamic agents in a surrounding environment of an ego vehicle. In one embodiment, a method includes, in response to receiving sensor data including present observations of a road agent of the dynamic agents in the surrounding environment, identifying previous observations of the road agent from an electronic data store. The method includes estimating a future state of the road agent using at least the present observations and the previous observations of the road agent to compute the future state according to a probabilistic model comprised of a transition model that accounts for dynamic behaviors of the road agent to characterize transitions between states, and an agent model that accounts for actions of the road agent. The method includes controlling one or more vehicle systems of the ego vehicle according to the future state of the road agent.

Systems and methods for agent tracking

System, methods, and other embodiments described herein relate to improving situational awareness of a user. In one embodiment, a method includes, in response to acquiring, in a personal device, sensor data from a set of sensors that perceive a surrounding environment of the user, identifying from the sensor data one or more objects in the surrounding environment relative to the user. The method includes determining a threat to the user from the one or more objects according to at least trajectories of the one or more objects embodied in the sensor data. The method includes generating, via the personal device, an indicator to inform the user of the one or more objects according to the threat.

Systems and methods for improving situational awareness of a user

The technology can advantageously improve the speed of software verification using an approximation of a datastore storing a multiplicity of indexed data. An example method may determine data point(s), suspected to violate a specification of a software program being verified, using approximation(s) of datastore(s) of the software program. Data ranges reflecting subset(s) of data from the approximation(s) that contain the data points may be determined and data entries stored in the datastore(s) may be searched for the data point(s) suspected to violate the specification. The data entries lie within the data range(s) and the method determines whether the data point(s) are in violation of the specification based on the searching.

Falsification of software program with datastore(s)

The technology can compute an approximation of a datastore storing a multiplicity of indexed data. An example method can a template including programming logic that, when executed, calculates output(s) based on input(s) and undetermined parameter(s). The undetermined parameter(s) are input into a machine learning framework. Data entries reflecting one or more inputs are retrieved from a datastore and input into the machine learning framework, which determines value(s) for the undetermined parameter(s), respectively (making them determined parameters). The example method generates an approximation of the datastore using the determined parameter(s) and the input(s).

Growing community of inventors

San Mateo, CA, United States of America

Nikos Arechiga Gonzalez